Consistent crashes after update to 4.6.2

Install, Configure and Update
1

I recently updated to cryoSPARC v4.6.2 from v4.1.2.

Following the update cryoSPARC is able to start and typically run for only a few minutes down to just a couple seconds before it crashes the web app. Running cryosparcm status after the crash I see the following:

CryoSPARC process status:

app STOPPED Not started
app_api STOPPED Not started
app_api_dev STOPPED Not started
command_core RUNNING pid 80112, uptime 0:07:11
command_rtp STOPPED Not started
command_vis STOPPED Not started
database EXITED Feb 28 10:18 AM

It does not correct itself when I run cryosparcm restart, but I can get back up and running for a bit when I restart the app, app_api, and database separately. But ultimately it returns to the same status as above in seconds-minutes.

Other observations:

  1. presence of *.lock files in the database suggest a database error?
  2. when the GUI is running, it shows that I am still running the old version of cryosparc and suggests an update despite the terminal showing the updated version.
  3. Immediatly following the update cryosparc wouldnt launch because of low diskspace - this was resolved by just emptying the trash folder on the disk. Not sure if low diskspace prevented something from completing the update.

any and all assistance would be great. I have a project attached to this computer that at a minimum I would like to be able to detach and continue processing on a separate workstation while troubleshooting this one.

Best,
Justin

2

Update: I made my database a backup and reinstalled the database and this has fixed the crashing issue. I am still not certain what the initial cause was. However, under the new database installation I have lost all user accounts - is it possible to transfer the user accounts and attached projects into the new database?

3

Further update: I have been able to create a user account and attached the detached project. CryoSPARC and the GUI appear to be working smoothly. I can access projects and view the data and logs of previous jobs but it is not allowing me to Queue any newly built jobs. It seems there is now a disconnect with the hardware. If i go to “manage → instance” I do not see anything on this tab in the GUI.

4

If the CryoSPARC database directory, whose location/path can be displayed with the command

cryosparcm status | grep CRYOSPARC_DB_PATH

was also stored on the full volume, the CryoSPARC database may have become corrupted. It is important to never allow that volume to run out of space.

The CryoSPARC database stores information about UI users, projects and GPU worker resources. I suspect that after re-installation of CryoSPARC, $CRYOSPARC_DB_PATH pointed to a blank database. You have already created a user and attached a project directory:

but may not yet have connected a worker.
To enable us to suggest further steps, please can you let us know:

  1. Is/was your CryoSPARC installation of type single workstation, master-worker or cluster?
  2. What is the output of the command
    cryosparcm status | grep -v LICENSE_ID
  3. If you want to cache particles, what is the location/path of the relevant directory?